As a designer of a utility that interprets media file formats, I am certainly in full agreement. Codec designers really should register the new ftyp codes they invent (hello.. Nero...! ;)   As David says, it's free and only involves sending an email. Here's the direct link to the details of submitting the request.

It defeats the purpose of following an ISO or other standard when doing so allows a file parser to get right to the field of interest... only to find there's no way of interpreting the contents it's arrived there.

And there are other problems with non-registration. Though it's a bit unlikely, it's entirely possible that, without an accurate central repository, two vendors may end up inadvertently using the same code to mean two completely different things. That, of course, makes the contents of the field entirely meaningless. And not just for information utilities like GSpot. It would likely have an adverse impact on both vendors as well, even if their original thought was that ftyp code was for internal use only. And I would argue that such a mix-up would arguably be the "fault" of the first vendor; i.e. the one who didn't take proper "ownership" of the code in the first place.

I would assume that vendors like to see their files identified so precisely by third party utilities, as seen in these new GSpot screenshot excerpts. In this case I was able to achieve this by tediously collecting information from field samples, hearsay information, experimentation, etc. But I prefer to display information based only on actual specifications and their associated "registration authorities". It's clearly more accurate, and I certainly don't have the time to track down large quantities of undocumented information. If nothing else, registration is a simple way for vendors to ensure that their latest codes are included, and included accurately, in utilities such as GSpot. I believe this is a benefit to everyone involved , quite possibly including "bottom line" advantage to the vendor.

- Steve